Lexicon for TTS in AP

Hello,

Me and my colleagues would love AP’s TTS (Text to Speech) to have a custom definable lexicon feature. By lexicon I mean a file allowing users to specify how certain text formations (words) should be interpreted by the TTS feature. Basically I am joining the previous request mentioned here: https://talk.atomisystems.com/t/applying-amazon-polly-lexicons/5205 I would just need it to be working for the MS Azure voices as well.

Reasoning behind the request:
It often happens that the company’s name, name of the product, or certain abbreviation is quite unordinary which means that it is not red accordingly by the TTS software. This causes a need for manual editing every time such word is used in the captions which is unpleasant, tedious and increases the chance of human caused error in the project. Trust me that it is really embarrassing when the name of the company you work for is misspelled in your own video ;-).

In real-life work scenario after I finish making a video in AP I send a ClosedCaptions(.srt) file together with the video to a language check to an English native speaker. They send me back the .srt file with a suggestion for changes I either accept or decline the suggestions, then import the .srt file back into the AP do an Batch Operation to Convert CC to Audio. After that I need to go through the whole project to edit the pronunciation of specific words and sentences manually by adding SSML tags like this:

<speak version=“1.0” xmlns=“W3C Speech Synthesis namespace” xml:lang=“en-US”>
<voice name=“en-US-JennyNeural”>
<phoneme alphabet=“ipa” ph=“təˈmeɪtoʊ”> tomato </phoneme>
<
/voice>
<
/speak>

At this point, when the SSML tags are in place, I can no longer make any extensive changes to the project, because exporting and importing the.srt file (sending it for language check) would overwrite the content of the SSML with the caption text.

Those are the reasons why I strongly believe that the Lexicon feature is necessary for any software using TTS and why I and my colleagues would love to have it in the ActivePresenter as well.

Here is some info on the lexicon and its functionality on the MS Azure side:

https://docs.microsoft.com/cs-cz/azure/cognitive-services/speech-service/speech-synthesis-markup?tabs=csharp#use-custom-lexicon-to-improve-pronunciation

Hi Vit,

Thank you for your suggestion and detailed explanation. We will try adding more options for TTS in ActivePresenter (volume, speed, pitch, lexicon…) in future releases.

Regards,